sate_ai_cli library

CLI-safe library for SATE AI — no Flutter SDK dependency.

Use this import in Dart CLI executables (bin/) or pure-Dart tools that must run with dart run (i.e., without the Flutter toolchain).

Flutter-specific adapters (TFLiteAdapter) are intentionally excluded from this library. Import package:sate_ai/sate_ai.dart in your Flutter app code to access the full API including TFLite support.

Classes

AIInput
Text or binary input payload for an AI model inference call.
AIModelAdapter
Abstract interface that wraps any on-device AI model.
AIOutput
Output produced by an AI model inference call.
BaselineComparison
Result of comparing a report against a baseline.
BaselineManager
Manages golden baseline comparisons for stress test reports.
BatchItem
A single batch item to run.
BatchProgress
Progress update for batch execution.
BatchReport
Complete batch report summarizing results across all models tested.
BatchResult
Result of a single batch item execution.
BatchRunner
Runs stress tests on multiple models in batch mode.
BenchmarkReport
Performance benchmark report for AI model inference.
ConfidenceThresholdInjector
Validates that the model's output confidence stays above a threshold.
Failure
Records an unexpected exception that prevented a fault cycle from running.
FaultInjector
Abstract base class for all SATE AI fault injectors.
FaultResult
The result of a single fault injection + inference cycle.
FinishedEvent
Event emitted when the test finishes.
InjectorCompleteEvent
Event emitted when an injector completes.
InjectorErrorEvent
Event emitted when an injector errors.
InjectorStartingEvent
Event emitted when an injector starts.
LatencyInjector
Simulates increasing inference latency over repeated calls.
LogEvent
Event emitted for log messages.
MalformedInputInjector
Injects malformed, boundary-violating, and random inputs to test that an AI model handles bad data without crashing or producing corrupt output.
MemoryPressureInjector
Simulates memory pressure to test how an AI model behaves near or beyond its available RAM budget.
MetricDeviation
Represents a deviation in a metric during baseline comparison.
MetricDiff
Represents a difference between two reports for a single metric.
MockAdapter
A fully-in-process mock implementation of AIModelAdapter for testing.
ModelSwapInjector
Simulates model corruption or swapping to a degraded model version.
QuantizationDriftInjector
Simulates gradual precision loss in model outputs.
ReportComparator
Compares two stress reports and generates a diff.
ReportComparison
Result of comparing two reports.
ReportDiff
Result of comparing two stress reports.
SateAI
Top-level convenience API for running SATE AI stress tests (CLI edition).
StartedEvent
Event emitted when the test starts.
StressEvent
Base class for all stress test events.
StressReport
The aggregated result of a full SATE AI stress test run.
StressRunner
Orchestrates a sequence of FaultInjector runs against an AIModelAdapter.
StressScheduler
Scheduler for running stress tests at specified intervals.
ThermalThrottleInjector
Simulates CPU thermal throttling under sustained load.

Enums

DiffStatus
Status of a metric difference.
FaultType
Types of faults that can be injected into AI model stress tests.
MalformedInputKind
The variants of malformed input that MalformedInputInjector generates.
StressEventType
Events emitted during stress test execution.

Extensions

FaultTypeX on FaultType
Extension helpers on FaultType.

Typedefs

EventCallback = void Function(StressEvent event)
Type alias for event stream callbacks.

Exceptions / Errors

AIInferenceError
Thrown when an AI model inference call fails.